Tuesday January 31, 2006 NeoOffice 1.2 Released
This just in: Patrick's satisfied with the beta code and has released NeoOffice 1.2 - you can download it now if you're a Mac OS X user on PowerPC.
NeoOffice Tipping Point
I was chatting with Patrick Luby at the weekend and thought his story was worth telling here. Patrick is the main developer of NeoOffice, the other port of OpenOffice.org to the Macintosh OS X platform. I've been using NeoOffice ever since I switched to the Mac because it doesn't need X11 like the "official" port and because it it well integrated with OS X - there's font, printer, drag and drop and other integration, making the experience very close to any other OS X application. Patrick works by taking the "official" port and using his own unique mix of native and Java code to produce NeoOffice. Using the clause in the LGPL that allows it he licenses the whole thing under the GPL.
For quite some time now, Patrick has been dependent on donations from grateful NeoOffice users to keep the wolf from the door while he does the porting. At one point the donations were streaming in, allowing him to work full-time on NeoOffice and not have to take other contracts. But according to Patrick, publicity about the work of the "official" porting team around the time of OOoCon claiming they would achieve a full native port and ditch X11 led one or two key donors to believe Patrick's work was no longer needed, and as the end of the year approached, donations faded away.
Patrick went looking for a permanent job, and received some offers. He was about to accept one of them when he spotted that, in the Terms of Employment, he saw he could potentially be required to assign his copyrights to any personal programming project - like NeoOffice - over to the new employer. While the employer tried to assure Patrick that that would never happen, the risk was enough to make him wait just a little longer - who'd want to work in a job that could end up stealing your baby?
As 2006 has got started, the flow of contributions has picked up a little and an 'angel' donor has made a gift to him so he's decided to put off a full-time job for a while longer and get started on the port of OpenOffice.org 2.0 over to the NeoOffice framework, especially seeing as the X11 port of 2.0 is looking in good shape even on Intel (great work, team!). He plans to create both Intel and PowerPC versions. He'll start that once he has a full release of NeoOffice 1.2 finished (that's based on the last version of OO.o 1.1.5, with read-only support for ODF), which I gather could be really soon.
I thought those of you who, like me, depend on NeoOffice would want to know and perhaps consider tossing a few notes in Patrick's tip-jar so we can be the "tipping point" in all senses of the phrase. Free software developers have to live, after all.
Monday January 30, 2006 links for 2006-01-31
-
Memo to dictatorships: the Internet is designed to route around you.
Strong Sun Showing in Developer.Com Awards
I notice that developer.com has just completed its annual awards event. I was especially pleased to read:
Again this year, Sun turned out to be the company taking home the most wins. Sun repeated their honors by taking home five of this year's awards. Java related products overall turned out to be a big winner.
The Sun winners make an impressive list:
- Framework of the Year
- Wireless/Mobile Development Tool or Add-in of the Year
- Java Tool/Add-in of the Year
- Security Tool or Add-in of the Year
- Framework of the Year
And like the Editor, I was surprised and delighted to see that the open source tool of the year was OpenOffice.org 2.0 (jointly with Firefox) - a random guess would have assumed that Eclipse would win that category, if only for the sake of political correctness. Both the joint winners are projects in which Sun invests, so that's an extra pleasure.
Overall, that result is a great indicator to me - open distribution models, platform-netural tools and open source communities are gaining ground rapidly. If any of this is a surprise to you I suggest you pop over to the Sun Developer Network and join right now (it's free, of course)...
Sunday January 29, 2006 Dick and Ron in line for a Jolt?
This just in: Ron Goldman tells me the book he and Richard Gabriel wrote has been shortlisted for a Jolt Award. The book, Innovation Happens Elsewhere, is a textbook for commercial software developers who want to go open source, and the authors have been instrumental in the introduction of open source methods throughout Sun over the last decade.
Huge congratulations to them both on the recognition - I hope they win when the actual awards are announced on March 15.
links for 2006-01-30
-
"The greatest enemy of freedom is a happy slave".
-
Roumen's worked it out! Great work, I know lots of people who want to do this.
-
Red Herring profiles the Eclipse project. I'm quoted in connection with NetBeans and I stand by that position; what is the sense of paying a vast sum ($250,000) to join an organisation that doesn't make any software we use?
Open Media Commons Workshop
I notice that the Open Media Commons (the initiative to create open and open source DRM) has announced a workshop in California on March 15-16. Looks worth investigating and registering early - and they are inviting presentation proposals too.
Saturday January 28, 2006 links for 2006-01-29
-
See what direct experience does for understanding. Maybe we should start a fund to buy every Senator an iPod and a $15 iTunes credit?
-
Still secrets so close to our largest cities. It's a wonderful world, I hope this discovery doesn't end that.
-
When "fair use" becomes "customary historic use" then innovation becomes the sole domain of monopolists.
-
Sun's President Jonathan Schwartz reflects on the possibility of dual-licensing OpenSolaris.
DRM and the Death of a Culture
I had the privilege of delivering a keynote at the Open Source Meets Business conference in Nürnberg, Germany this week (delegates will find my slides online as a PDF). I travelled there from an engagement in Paris, and took the Metro/U-bahn2 in both places. There was a very visible difference between the two experiences.
In Paris, I bought my Metro ticket and then used it in an automated barrier to reach the platform. I noticed lowlife furtively scanning the station and then vaulting the barriers, and I saw armed police at the station to catch the thieves doing this (they didn't catch any that I saw, and there were several of each at each station).
By contrast, the U-Bahn in Nürnberg has no barriers. I bought my ticket, boarded the train without fuss, there was no risk of being shot by a policeman targeting a barrier-vaulting cheat, and the system was still clean, efficient and well-used.
This all sprang to mind when a conversation about DRM followed the GPLv3 item up over on Stephen's blog. A comment writer (Christopher Baus) said of DRM:
I might be the only technologist on the other side of the DRM fence. To me it is like checking my lift ticket when I get on the ski lift. I might find that a bit annoying, but if ensures the resort can stay in business from collecting ticket money, then that is a net good thing for me. If the ski resort goes out of business I can't go skiing, and I would resent those who got on the lift w/out paying.
I think there are quite a few people around who have Christopher's view, which is unfortunately rather simplistic. DRM - the imposition of restrictions on usage of content by technical means - is far more than that. It's like checking the lift ticket, yes, but also the guy checks you are only wearing gear hired from the resort shop, skis with you down the slope and trips you if you try any manoeuvers that weren't taught to you by the resort ski instructor; then as you go down the slope he pushes you away from the moguls because those are a premium feature and finally you get to run the gauntlet of armed security guards at the bottom of the slope checking for people who haven't paid.
DRM's Collateral Damage
The problem with technology-enforced restrictions isn't that they allow legitimate enforcement of rights; it's the collateral damage they cause in the process. In my personal opinion the problems are (very concisely) that they:
- quantise and prejudge discretion,
- reduce "fair use" to "historic use",
- empower a hierarchical agent to remain in the control loop, and
- condemn content to become inaccessible.
To go into more depth on those:
- Technology-enforced restrictions quantise and prejudge discretion
- People talk of "fair use" but what they actually mean is that we all depend on the exercise of judgement in every decision. Near the "bulls-eye" of copyright we're all clear what is what, but as Lessig eloquently explains in Free Culture, in the outer circles we have to make case-by-case judgements about what usage is fair and what usage is abuse. When a technologist embodies their or their employer's view of what's fair into a technology-enforced restriction, any potential for the exercise of discretion is turned from a scale to a step and freedom is quantised.
- Technology-enforced restrictions reduce "fair use" to "historic use"
- The natural consequence of having the quantum outlook and business model of one person replace the spectrum of discretion is that scope for new interpretations of what's fair usage in the future is removed. Future uses of the content involved are reduced to just historic uses the content had at the time it was locked up in the DRM wrapper. The law may change, the outlook of society may mature but the freedom to use that content according to the new view will never emerge from the quantised state the wrapper imposes. The code becomes the law, as Lessig again explains in Code. As others have pointed out, "fair use" is forward-looking, "historic use" is ossifying.
- Technology-enforced restrictions potentially empower a hierarchical agent to remain in the control loop
- When use of content depends on a technology from a single vendor, as is currently universally the case, that vendor effectively becomes the gate to ongoing use rather than the actual copyright owner. A great argument for an alternative approach like the open source, distributed-identity-based scheme in Project DReaM. What's much worse, though, is that the restrictions don't go away when the rights they are enforcing do. Copyright eventually runs out, but technology-enforced restrictions never do.
- Technology-enforced restrictions condemn content to become inaccessible
-
As DRM's outspoken critic Cory Doctorow points
out, DRM condemns content to suffer the fate which for documents I
call corporate
Alzheimer's. Each of the problems above combine in a 'perfect
storm' to create a content owner's dream world of built-in
obsolescence and repeated opportunity to sell the same content to the
same people all over again if they actually like it enough to use it.
Meanwhile, our collective cultural memory gets locked up in instances
which become inaccessible on the occurence of the first one of:
- The content being 'turned off' by a usage rule
- The implementation of the restriction mechanism is obsoleted by an "upgrade" of the host system
- The original medium degrades into uselessness but couldn't be copied
- A part of the "phone home for authorisation" chain goes out of business
- The original license is no longer applicable (for example because your children have inherited the media but not the digital key)
Complacency leads to servitude
Christopher went on to express his complacency over the whole situation:
IMHO if DRM isn't good business it will go away. Simple as that. No need to worry.
Except the prior market power of huge corporations is being used to project it into markets in a way that distorts market forces and conceals the lack of ethics and the erosion of the social contract behind rights law. What if it's good business but bad humanity? What we're seeing here is the 21st century equivalent of enclosure - indeed, the comment by More from Utopia rings eerily true:
... not contented with the old rents which their farms yielded, nor thinking it enough that they, living at their ease, do no good to the public, resolve to do it hurt instead of good. They stop the course of agriculture, destroying houses and towns, reserving only the churches, and enclose grounds that they may lodge their sheep in them.
It's clearly right to "pay the labourer a wage" but is that enough excuse to also condemn culture into the memory hole and enforce an economy of constant repayment for the same stuff? Is there a solution?
If there is, it will surely involve a fundamental rethink of rights legislation - patents and copyrights - that goes back to the social contract on which both are based3, giving limited and temporary one-time rights in exchange for the enrichment of society. We've forgotten that was the root of the whole system, and corporations now have the sort of entitlement culture they deride in individuals.
We also need the invention of schemes like the Open Media Commons that allow the technological equivalent of the Nürnberg U-Bahn for content and OpenDocument that guarantees future access to today's documents. And we need to recognise the point at which schemes like iTunes finally funnel us away from circumventable almost-locks into real servitude and not give in to the intentional seduction4.
But whatever the answer, we need it soon, because we're rushing headlong into a world that will be doomed to forget its culture and history - if it doesn't keep paying the protection money. As a card on my wall reminds me, "the biggest enemy of freedom is a happy slave".
- The title is a reference to Rookmaaker's influential religious book discussing how modern art signals an inner decay of society.
- That's the subway/tube/underground railway - interesting how even in English there's no agreement what to call it. Just to complicate matters further, it was actually the RER that I took...
- As I have commented on Groklaw, in the US Constitution you'll find this social contract in section 8, clause 8. The core assertion in my article is that the beneficiaries of those monopolies have become addicted to them and now want to make the monopolies permanent without amending the Constitution (or equivalent elsewhere). Payment for creative work is still justified; requiring payment in no way necessitates junking the constitutional limitation on the monopoly of rights. But the sole focus of the technology industry so far has been to do just that, at the behest of customers who want to make temporary monopolies permenenat.
- James Governor has been hammering me on this one for a while. I'm happy enough to use iTunes all the time there's a way to get real MP3s of my purchases that's not too much fuss as it also makes me take backups. But I'm pretty close to giving up on them because of their "5 users" limit (as a household we have 5 machines so we have to keep a careful eye on authorisations), and I don't buy their videos because I can't get unencumbered versions that will live on when iTunes dies.
Friday January 27, 2006 links for 2006-01-28
-
The perfect comment on google.cn
-
"So I don't think the GPL v3 conversion is going to happen for the kernel, since I personally don't want to convert any of my code."
-
Music label underwrites a fan's defence of RIAA's arbitrary judgement against them. "Suing music fans is not the solution, it's the problem," stated Terry McBride, C.E.O of Nettwerk Music Group.
-
Answers those questions about US national security and sets your worried head at rest.
-
Do Not Play This. It will steal all your time and destroy your life (it's a super-cool AJAX game of guess-the-tag).
Monday January 23, 2006 Open Source ahead of Microsoft
Quick note: It may not be finished yet, but there's already an alpha version of OpenOffice.org for the Intel-based Apple Mac. Kudos to Eric, Daniel, Stephan and Florian.
That's the beauty of open source - people can just get on with it, no need to ask permission. I've no idea when MS Office for Intel Mac will be ready, but I'm pretty sure that OpenOffice.org will be a great choice even when it stops being the only choice, not least because it supports the open standard file format that everyone (except "NIH" Microsoft) is supporting!
Sunday January 22, 2006 SDForum Fixings
I've had some interesting correspondence lately. It seems that in Dan Farber's blog about Sun and open source last week (connected with SDForum's event on the future of commercial open source), there are a couple of things that need adjustment. Firstly, he quotes me saying:
"If you believe in an open source business model, you are mistaken. Open source is tactic within a business model"
One correspondent took me to task, asking why I didn't believe in open source. That's the problem with reported speech; the rest of the context gets lost. Dan had a good try at conveying the context in his following paragraph (it looked fine to me), but what's perhaps lost is that the event had as its audience a group of people who all totally understand what "open source business model" might mean - a point neglected in The Register's coverage too. Facing me in the crowd at the SD Forum event were Ward Cunningham, Stephe Walli, Matt Asay and who knows who else. They did not need lots of detailed and/or politically correct preamble.
At that point in the conversation, the question was not "can a business get to use open source effectively" - everyone present already knew the answer to that. It was "is open source itself the sole distinguishing feature of a business". In that context, my answer was that a business has many different aspects to it and within that larger scheme of things the open source element is clearly one of a number of tactics for almost every business that uses it.
"Tactic" there is not derogatory. There's no single "open source business model" - rather, an array of options for including open source software as an element (a "tactic") in a business model. Maybe every such model then becomes an "open source business model"? I'll write more about this subject soon.
OpenSolaris for PowerPC
The other thing Dan picked up was my comments about how open source allows an ecosystem to build around a "long tail" of some kind - in this case for a Unix operating system - expanding the opportunity for everyone in the ecosystem without the need for any particular company to be "in control" - Sun certainly is not orchestrating the various elements of the OpenSolaris ecosystem, for example. Dan asked for me to identify some other companies in the OpenSolaris ecosystem as he thought it was all about Sun. I mentioned Nexenta and Genesi - both working independently from Sun - and Dan wrote about them briefly.
Unfortunately, he was a bit concise in his mention of Genesi and I have had correspondence asking me to point out that while Genesi are a sponsor of the effort, the community at Blastwave that's working on Polaris is a grass-roots effort. There's plenty of grass-roots work going on in OpenSolaris, like Polaris and the Nexenta port of Ubuntu, and folk who think it's "all Sun" are really missing a lot of the goodness out there.
Wednesday January 18, 2006 Start the stopwatch
OK, start the clock - let's see how long it takes The Register to acknowledge and correct the egregious error they made in the article Software patents loom large again. In that article, the anonymous writer asserts that
The bill had been supported by the European pro-patent lobby, which included corporations such as Microsoft and Sun, who claimed that the directive would encourage investment in research and development in Europe.
What rubbish! Microsoft was indeed a prominent proponent of software patents. On the other hand, as was well documented, Sun and Red Hat were the core of a small but significant industry group lobbying around the position that the directive would be harmful because it threatened the freedom to create interoperable free/open source software. Anyone who had been paying the slightest attention to the issue, or did more than the most cursory research, would know this was the case.
[Click! Corrected at 2pm PST Jan 18. Thanks, Drew.]
Tuesday January 17, 2006 Spreading GPL across SOA and AJAX?
Just back to my hotel after the second (final) day of the GPL v3 conference. I attended the Q & A panel (pictured to the right) in the morning, discussion committee B around lunch time (representing Sun) and discussion group A this afternoon (representing OpenSolaris). Today has been much more about getting doen to the gritty work of adding comments and discussion to the draft. Indeed, group A added a number of comments this afternoon and the comment flow is roaring ahead (fortunately there's a syndication feed).
I notice that Stephen O'Grady at Redmonk thinks like Mike Olson that the new GPL funked out on the provision to rein in ASPs. At first I thought so too, when I saw that there was no statement extending the need for publication of source for web execution of GPLed code and that it seemed to be a goal of section 7:
Aside from additional permissions, your terms may add limited kinds of additional requirements on your added parts
I think it's clear that the intent is indeed to initiate a marketplace for extensions. But as I have looked at it more and more, I believe Eben and Richard have been far more subtle. A crude and explicit ASP clawback would have raised a riot. Instead, the seeds are sowed in section 5c:
If the modified work has interactive user interfaces, each must include a convenient feature that displays an appropriate copyright notice, and tells the user that there is no warranty for the program (or that you provide a warranty), that users may redistribute the modified work under these conditions, and how to view a copy of this License together with the central list (if any) of other terms in accord with section 7.
and even more in section 1, "Complete Corresponding Source Code":
For example, this includes any scripts used to control those activities, and any shared libraries and dynamically linked subprograms that the work is designed to require, such as by intimate data communication or control flow between those subprograms and other parts of the work, and interface definition files associated with the program source files.
Doesn't that latter language just yell "SOA" to you? It seems to me that there is the germ of an idea here to ensure that, if your server software is licensed under GPL v3, the client software - AJAX, plug-in or whatever - is required by this language to also use GPL v3. This is a much more subtle approach, and strikes right to the heart of the rapidly evolving "read-write web".
By the way, if you're looking for a properly formatted version of the GPL v3 draft, Tim has created one. [Update: He's added paragraph-level links so I have linked sections above]
Monday January 16, 2006 GPL v3 Released
Using the miracle that is SubEthaEdit, Danese Cooper and I collaborated to produce the notes below during this morning's plenary session at the GPL v3 conference here in Cambridge, MA during the launch of the draft of GPL v3. Apologies for any errors - let one of us know what we got wrong.
Update: Fixed a number of typos and non-critical stuff in the Q and A.
Notes on Eben Moglen's presentation discussing GPL v3
- Instead of distribution, used "to propagate" - trigger not in any existing copyright statute.
- New language to tackle DRM - complete corresponding sourcecode also includes any encryption or authorization codes necessary to install or execute the source code of the work, Eben calls this "plays all the same movies"
- New "system library exception" - handled in a "special restriction" - aimed to clarify that many previously disputed combinations of code were actually within the spirit of the GPL.
- Patent Peace - no private use if you've started patent infringement against anyone within the scope of the license.
- Manifesto down to two paragraphs - includes statements on loss of freedom which may suffer from further interpretation...
- Language to assert that no GPL program can be a "circumvention device" for any other GPL program, by definition. Intent to ensure that the world of GPL cannot be impacted by DMCA & equivalents.
- Emphasised that the new wording seeks to ensure that there is no new exclusion of any "legitimate member of the community"
- Language covering embedded notices (copyright status, attribution, right of warranty). Portions of the work can carry additional terms. Each user interface must include a means to display copyright & warranty statements. One place to set those notifications for "machine readable" methods.
- "Separate works": although they stand separately under other licenses, may be distributed as part of a larger GPL whole. This is a clarity change not substantive change according to Eben.
- Aggregate work concept allows use of software in compilations without triggering GPL by inclusion of extra language.
- New requirement forces physical media availability...'for as long as you offer spare parts or other customer support". Still the option to provide online source as long as it is offered in the same place as the binary. (main change was extension of 3 year obligation)
- Obfuscation not allowed
- Add your terms section - permissions additional to GPL should be "in writing".
-
Compatibilities - Apache and others: limited types of additional
requirements may be added (add-on clauses permissible)
- to preserve certain notices or author attributions
- disclaimer of warrunty (trying to reduce national licenses)
- limit use of trademarks and specific names use
- allow users to obtain copies of code
- allow additional defensive patent retaliation - ASL and EPL meet this standard according to Eben.
- Removal of notices: remote services or patent requirements must be on a single aggregated list "unless the work permits distribution under a previous version of this license"
- 60 days notification prior to termination...auto-termination gone (justification in prior time was "not enough staff"). Now if you cure a "violation" promptly (60 days) and get no additional notification you can consider yourself "cured". FSF will continue to do compliance.
- Not required to accept the license in order to receive a copy of the program (no contract is created, still) 183 contract law schemes in the world...can't standardize globally...RMS statement against global copyright system. "We use it because its there. BERN + WTO not a good thing."
- "You are not responsible for enforcing compliance"
- When you distribute a covered work, you grant a non-exclusive, worldwide and royalty free patent license (no essential claims language).
- Prescribes what to do to prevent distributors from putting their downstream customers in danger from which they themselves are exempt. You must cascade protection you have to those receiving source from you. Request for help from patent holders and those who care about patents to work on this together.
- "Patrick Henry provision": If you can't distribute legally due to external conditions then you can't do so at all. G-Streamer plug-in under X11 license + patents and others (sum effect making it non-free). RMS calls the "Liberty or Death" and Eben calls it "Truth in Labeling".
- Geographical restrictions clause to go, rarely used.
- Revision clause of GPL...up to copyright holder to decide w/ each version (btw, the software written by Bradley Kuhn's team at SFLC and used to support public comments is licensed under GPLv3, so it will be the first software to use the license). SFLC will interact with every project that requests information. Not coercing anyone to pick it up. RMS says projects using earlier GPLs should "take steps to make sure they are in a position to move to GPLv3".
- Warranty exclusions not in all uppercase anymore. Added separate liability disclaimer for safety critical systems...can add warranty for a fee.
- Hoping to "shorten" the license, length of this preso notwithstanding ;-)
- Keen to avoid language that deals with disputes that E & R regarded as unnecessary.
- In the business of giving permission, not constraining outcomes...glad its up to the GPL community now :-)
RMS closing remarks - United States not optimal venue for some participants so apologizing...
Notes from the subsequent audience Q&A session
- Chris DiBona:
- Safety clause...are they allowing additional clauses? Possible RTOS problems.
- Eben:
- No, just outlining risk and need to pay for separate warranty. Andy Wilson: Add-on clauses (compilers for example) - are additional permissions such as those used by Classpath permitted? Eben: yes, they are additional permissions in the scope of section 7.
- ??1:
- Intent behind patent protection language? IBM and Microsoft: Microsoft recieves or authors GPL code and distributes, knowing that it infringes on IBM patent...does Microsoft need to protect me from downstream patent risk?
- Eben:
- Yes..if you don't have sublicense to protect downstream sublicensees, then you must take other precautions
- ??1:
- Knowledge (of the existence of patent issue) is the key?
- Eben:
- Yes...we are not asking people to be more diligent in discovery.
- ??1:
- In the event of patent claim unrelated to distributed work?
- Eben:
- Not involved in this. We will not abate the patent problem until everybody realizes we are all in it together.
- Henry Poole:
- 60 days to cure? What if they were violating 120 days ago?
- Eben:
- The goal is not to "catch" people who were violating for 5 days, 80 days ago, but have now cured the violation. If you didn't complain within 60 days then they are off the hook if they've cured...the goal is not to give a short statute of limitations, its to close the set of possible claimants. Most problematic situation: someone is distributing a large collection of free software packages and they fix it once we contact them? How do we help them now that they are liable to potentially hundreds of claimants? We used to help them round up all possible claimants but it wasn't workable. Now that class will be closed quickly. Don't believe it will hobble enforcement.
- ??2:
- Compatibility provisions are a good thing. Which licenses remain incompatible? Old BSD or OpenSSL?
- RMS:
- BSD Advertising clause not allowed. Don't want to make GPL compatible with things we are campaigning against.
- Eben:
- Goal of permission side is to take existing permissive Free Software licenses and make them compatible. Repugnant terms still not accepted, and there are still a bunch of existing licenses with patent retaliation clauses not compatible with GPLv3. Overbroad retaliation will still be a problem.
- Jeremy Allison:
- [Call for round of applause as there was no opportunity after the session. Audience responds warmly] DRM provision: with a proprietary chip that runs a particular signed version of GPL'd code...how does the new version compat that?
- Eben:
- Take an OS (the Hurd) and say you sell boxes that run Hurd from boot if and only if it is a signed copy. Signature codes will have to ship with your distribution. Tommy in his bedroom will be able to sign, boot and run or the distributor will be in violation.
- ??3:
- Promise to protect (warranty)...what is the effect on small developers?
- Eben:
- If you were relying on a patent license (not normally affecting small developers, but more large businesses). This is spoken to about 8 gentlemen in this room...they know who they are :-) Need to get a little more skin in the patent game.
- Yidong Chong:
- v2.2 project that hasn't aggregated copyright...do they need to get permission?
- Eben:
- Each distributor gets to make decision. Up to legal structure of project. SFLC is here to answer those questions.
- RMS:
- Most projects have no legal structure, so this is a problem, especially if code is licensed as "GPL v2 only" and has not stated "or any later version".
- Yidong Chong:
- Can you delete the previous version of the license when you implement the new one?
- Chris DiBona:
- Has a number of developers who work on GCC. Interface with patent covered libraries language makes him worry that GCC changes might not be able to run on Windows or Solaris for instance. Morale difficulties.
- Eben:
- Which is why we changed the provisions...is it inappropriate to link GCC? v2.2 didn't make it clear (although Eben & Richard believed it was clear), FSF thinks it should be permitted explicitly, hence the clearer OS exception.
- Chris DiBona:
- Looked like it said if a patent license is required, then you can't link. When someone uses licensed interface (Bruce Perens: TCPA)
- Eben:
- There is a problem. Not acceptable to put patented enhancements in userland. Patent claims that read on general interfaces are not an issue...OpenSolaris example therefore not a problem, but there are some similar sitations that might be a problem.
- Warren Togami:
- What about cases where binaries are signed for authentication rather than for restriction of use?
- Eben:
- In general, if the user has control of the keys then you are okay...its when you take away control from the user that incompatibility happens.
- WT:
- So what if its a "one-time key"?
- Eben:
- Yes that would work as long as the user can control
- RMS:
- Please submit as a comment, it seems to need clarification.
